Calendar-based maintenance services equipment that does not need it and misses equipment that does. Condition-based monitoring replaces the calendar with evidence — but only works when the underlying data model is sound enough to tell a developing fault from a sensor artefact.

LAMR builds condition-based monitoring and anomaly detection on top of the asset context we establish first, because detection built on unmapped tags produces alarms nobody trusts and everybody eventually ignores.

Condition-based monitoring

  • Condition indicators per asset class. Defined once at template level so they apply across the fleet, not configured per machine.
  • Operating-state awareness. Thresholds that understand startup, load-following, standby and outage, so normal transitions stop generating alerts.
  • Degradation trending. Tracking the direction and rate of change rather than the instantaneous value, which is what gives maintenance planning useful lead time.
  • Maintenance integration. Linking condition findings to work orders so the loop actually closes.

Anomaly detection

  • Multivariate detection. Equipment rarely fails on one signal. Detecting a changed relationship between signals catches faults that threshold alarms never see.
  • Baselining against real history. Models trained on the archive, with known bad periods excluded rather than averaged in.
  • Quality-code awareness. A stale tag is not a stable process. Detection that cannot tell the difference generates confident nonsense.
  • Alert tuning. Deliberate management of false positives, because the failure mode of these systems is being switched off.

Asset performance management

Beyond detection, the same foundation supports fleet-wide performance comparison, remaining-useful-life estimation, efficiency and heat-rate tracking against design, and availability analysis that distinguishes genuine equipment loss from operational decisions.

Making it operational

An insight that never reaches an operator is a research project. We deliver detection into the tools your teams already use — PI Vision, existing dashboards, the alerting path the control room actually watches — with documented tuning and handover so your engineers can adjust it without us.